Along with antibodies, selected nucleic acid binding species ( aptamers ) are one of the few classes of molecules that can be selected to recognize a comprehensive set of protein targets. However, unlike antibodies, aptamers can be adapted to serve as reagentless biosensors. We have previously developed several methods for the production of so-called 'signaling aptamers' that can directly transduce molecular recognition into optical signals. However, none of these methods lend themselves to the high-throughput production of biosensors for proteome chips. Therefore, in Phase I of this application, we propose to develop and compare methods for the high-throughput production of signaling aptamers. While all of the methods that are advanced can potentially be generalized to numerous protein targets, we will initially focus on four relevant targets, bFGF, VegF, MEKl, and ERK2. Following the development of these methods, we will choose the best biosensors that have been produced and demonstrate that they can function in chip arrays. Overall. our criterion for success in Phase I is to generate a four-sector chi that can specifically detect the two cytokine and two kinase targets listed above in a physiologicallv relevant milieu. The method or methods that prove most successful for the generation of sensitive and specific biosensors will be adapted to a high-throughput, automated format in Phase II of the application. At the end of Phase II we will have generated a 1OO-sector chip that can specifically detect a range of signaling kinases and also distinguish and quantitate their modification states. This work can be immediately extended to full proteomes under the aegis of a national facility or company.